I'm sure it's a good article but there is no need for it to be written by an LLM... Is everyone really this lazy now? I don't think I've read anything written by a human for at least a full month.
Everyone should aim to be this lazy. Everyone is trying to maximize their time and now. LLMs got decent at writing, and before long they will be better at write-ups than the average human and we won't look back.
Of course, this is contingent on your current goal: If you want to get better at writing, you should do the writing, just like you want to do the math, if you want to learn multiplication. Writing is fine as a tool to get information across. It needn't be more than that for everyone.
But this is what software has always done. It helps people be lazy in some respect, and then, so far, we have always found new stuff to do with the time we freed up.
The need is that writing is expensive. An LLM is around a hundred times cheaper than a human, so this kind of article that doesn't provide tons of value just makes sense to write with an LLM.
That might apply to some things, but there is little to no value in this article, so the economics are the same as the spam - cheap to spew garbage while providing no value and wasting the readers time. It's not something that should be encouraged. RIPE should know better.
I've been running my own email server with exim and dovecot for 20ish years, I like the freedom it gives me but I have to say it's a significant investment. Not so much keeping things up to date but instead making sure that my emails are actually delivered. I've eventually switched to using fastmail just for sending emails and even with that, I don't get perfect delivery.
I wish there were a way to self-host without my emails constantly getting stuck in other people's spam folders. And it made me really paranoid about email, too -- like I'd send emails, and I'd never know if the other end received them or not, unless they replied. Without a reply, I would always wonder whether the message got a look, or disappeared without a trace.
The only way around this seems to be hosting with Google, Microsoft, or maybe Proton. Any other ideas?
> DMARC: adopted everywhere, enforced nowhere in particular
Unfortunately this only talks about what DMARC policies are published by domains. I was hoping they will go into some analysis of how many mailboxes actually hard-enforce DMARC rules on ingress (as in, respect the DMARC policy as published by the sender's domain, p=reject or whatever else it might be)
I have been hesitant to enforce DMARC rules on my server. Partly because maintaining it was a pain (see opendmarc - maybe better with rspamd now), but also because on my test runs I saw plenty of legitimate incoming mail that simply would get rejected if I were to enforce DMARC on ingress.
I suspect big players like gmail have some kind of soft enforcement/reputation/machine learning approach to DMARC, but this is way too much work for a small time admin like me.
If someone can solve the spam problem in email without just manually whitelisting specific providers that heavily monitor their users, then they can completely reverse the direction of this problem. Unfortunately it doesn't seem likely someone could do that.
That would work if most spam wasn't from the cartel. I receive single digit percentage of "spam" from randoms, 95%+ is from Gmail/Google, o365 or mailgun/whatever. Not one of them can either filter their users or actually action abuse reports.
I enforce strict spec and behaviour compliance, as well as policies (SPF, dkim, etc). Aside from a very small number of well known spam/abuse sources, it's largely from legitimate mail accounts on Gmail/outlook etc. or idiot msps.
I wish that was true for all cases. My uni email is 99.9% spam and everything comes from official accounts sending irrelevant stuff to everyone and their grandma
It's not in the current providers' interest to help people avoid the current providers' advertisement or avoid paying the providers to host the customers' email.
You mean filtering? That could get extremely expensive computationally, though. Tools like Rspamd support LLM-based filtering, but traditional Bayesian filtering is vastly cheaper and more computationally efficient. Hm, maybe LLMs can be used as an additional signal for ambiguous messages rather than as the primary filter.
If you are thinking using AI to classify messages, the scale of email spam is so massive that using conventional techniques won't work. It won't make sense energetically speaking IMO.
I think techniques like HashCash by Adam Back should be revisited.
The problem lies with the big e-mail providers marking your e-mail as spam, even if you have everything configured right (MX, SPF, DKIM, DMARC, etc.) and you're not on any blacklists. They use their mysterious opaque algo to determine that your e-mail is spam and you have no idea what you could do to not get in peoples junk boxes.
> They use their mysterious opaque algo to determine that your e-mail is spam and you have no idea what you could do to not get in peoples junk boxes.
And that’s not hard?
Running a modern mail server, a spam filter, a self-hosted webmail, and learning just enough about DNS and networking to get mail through: this is what’s not that hard.
Figuring out how to stay whitelisted, or accept ending up in people’s filters, asking them to whitelist you while your public IP gradually gains trust, is a job.
20 years of running the classic postfix + dovecot stack for about 10 business domains. I’ve recently been testing out stalwart, with just one company domain, which is nice but seems quite fragile, 4 times now it’s killed itself doing some self updates. If it wasn’t for Claude assistance I probably would have given up on it by now. It is quite feature rich though, but its spam detection is not great, too much ham flagged as spam, not enough spam flagged as spam. 2FA is a nice bonus but for some unknown reason it breaks itself every few weeks and I gotta turn it back off again. If I could get stalwart stable I would migrate all my domains to it.
The hardest thing in setting up your emails server is to keep your messages out of spam folders. Even if you are completely legit company that sends out ONLY human-written personal emails, you will get tagged as spam first.
I've self-hosted for about 2 months, and promptly switched to iCloud. Just because I've been spending too much time calling people and telling them to check their spam folders.
Recently our business moved to another domain. We had a guy who has spent over a month "warming up" traffic, so our new domain won't be tagged as spam by default.
And on the other side, projects like stalwart email ( http://stalw.art/ ) and mox ( http://xmox.nl/ ) makes it so much easier to build an e-mailsystem on your server.
Nothing will ever ruin email. A private email server, or a federation of mutually-whitelisting ones, is the ultimate social media, and not at all a bad personal information management system too.
powerful enough models are becoming a reality on personal hardware. Olares is an open-source personal AI cloud OS, supporting local AI, Ollama, Open WebUI, Hermes etc
Hosting my own server with mailinabox. low traffic, but keeping their random subscriptions, newsletters and occasional single purpose address. Haven't had any issues so far.
I switched to M365 as my provider a few years back as my emails kept being flagged as spam.
I've done it for decades, but I don't know much beyond the basics of email hosting tbh, however my understanding is that these days email coming from self-hosted servers carries a much higher spam/junk risk and so often automatically marked as spam.
Maybe there are ways to avoid this happening, but when a few clients all reported that they were not receiving my emails I didn't have much of a choice but to switch.
This very much depends you been U.S. citizen or not been U.S. citizen.
"PRISM is a code name for a program under which the United States National Security Agency (NSA) collects internet communications from various U.S. internet companies. The program is also known by the SIGAD US-984XN. PRISM collects stored internet communications based on demands made to internet companies such as Google and Apple under Section 702 of the FISA Amendments Act of 2008 to turn over any data that match court-approved search terms."
"Section 702 permits the Attorney General and the Director of National Intelligence to jointly authorize targeting of non-US persons reasonably believed to be located outside the United States.
By targeting, it is meant that US persons or persons located in the United States may not be the intended targets of the collection. The targeting must have as its object the acquisition of foreign intelligence information, as this is defined by FISA."
I remember when Gmail came around and everyone was more than happy to hand off their email to Google. No spam and ability to send and receive huge (for the time) attachments I think were the main things people wanted.
Not sure if that was before or after everyone was happy to hand off their web browser to Google.
The alternatives to Gmail, at the time, were either Hotmail, or email provided by your ISP who charged money for additional email accounts. Those were the options.
You like using Google and Microsoft products as an end user (even ignoring all the ethical issues with these companies)? That seems surprising to me. I find their products so horrible compared to e.g. Fastmail.
We have to figure out a way to get out of FB social, iOS/Android phone, Google/Microsoft email, AWS/GOOG/MSFT cloud oligopolies. There is no competition and there will be massive rentier capitalism extraction from all these players for the rest of civilization.
TBH, I blame IT leadership. The easy choice is Google or Microsoft. "No one ever got fired for buying IBM" is now "No one ever got fired for subbing to Office." Someone needs to bite the bullet and stitch together an alternative array of services. They've cultivated their own monoculture by hitting the easy-button and forking over for Office365, etc.
Which if we're being honest, Microsoft and Google products are rarely best-in-class at any one thing that they do. Teams, SharePoint, Google Docs, someone somewhere is doing those things much better. Those competitors just don't necessarily have a whole ecosystem that makes it a big Easy Button for corporate IT to smash. And they sometimes charge as much by themselves as Office365 costs in whole. But I know one problem my own employers IT team struggles with a bit is putting out the constant fire of different teams deciding the Microsoft tool we pay for sucks and going rogue and trying to subscribe to something else. Dropbox was a big one, but Teams and Planner alternatives come up a lot too.
Not pretending to have all the answers but it doesn't seem to be a completely intractable problem.
It starts by just avoiding them as much as possible. E.g. self-host email. If you encounter a system that refuses to receive your email, I give you permission to use your old Gmail account with that system, while using your self hosted account for everything you can.
If your email is Google or Microsoft, there are other people with self-hosted email who feel pressured to use their Google or Microsoft accounts to send email to you.
I'm sure it's a good article but there is no need for it to be written by an LLM... Is everyone really this lazy now? I don't think I've read anything written by a human for at least a full month.
> Is everyone really this lazy now?
Everyone should aim to be this lazy. Everyone is trying to maximize their time and now. LLMs got decent at writing, and before long they will be better at write-ups than the average human and we won't look back.
Of course, this is contingent on your current goal: If you want to get better at writing, you should do the writing, just like you want to do the math, if you want to learn multiplication. Writing is fine as a tool to get information across. It needn't be more than that for everyone.
But this is what software has always done. It helps people be lazy in some respect, and then, so far, we have always found new stuff to do with the time we freed up.
This seems like a load bearing issue. Would you like me to put together a plan to put human-written text in your daily routine?
Oh you.
Are we even sure the actual stuff in the article is real? What if it's just the usual LLM hallucinations
> The domains didn't disappear; they moved.
I stopped reading at this point. I cannot emphasize how much I HATE this AI writing style.
The need is that writing is expensive. An LLM is around a hundred times cheaper than a human, so this kind of article that doesn't provide tons of value just makes sense to write with an LLM.
I hate LLM writing, but the economics checks out.
That might apply to some things, but there is little to no value in this article, so the economics are the same as the spam - cheap to spew garbage while providing no value and wasting the readers time. It's not something that should be encouraged. RIPE should know better.
How about not writing the article at all then?
Because people will read it if you write it, hence us being here.
I've been running my own email server with exim and dovecot for 20ish years, I like the freedom it gives me but I have to say it's a significant investment. Not so much keeping things up to date but instead making sure that my emails are actually delivered. I've eventually switched to using fastmail just for sending emails and even with that, I don't get perfect delivery.
I wish there were a way to self-host without my emails constantly getting stuck in other people's spam folders. And it made me really paranoid about email, too -- like I'd send emails, and I'd never know if the other end received them or not, unless they replied. Without a reply, I would always wonder whether the message got a look, or disappeared without a trace.
The only way around this seems to be hosting with Google, Microsoft, or maybe Proton. Any other ideas?
I don’t self host, I pay zoho to do it. Over never had any suspicion my emails haven’t been received.
I think that is the Crux of the issue.
Email spam has made self hosting untenable.
> DMARC: adopted everywhere, enforced nowhere in particular
Unfortunately this only talks about what DMARC policies are published by domains. I was hoping they will go into some analysis of how many mailboxes actually hard-enforce DMARC rules on ingress (as in, respect the DMARC policy as published by the sender's domain, p=reject or whatever else it might be)
I have been hesitant to enforce DMARC rules on my server. Partly because maintaining it was a pain (see opendmarc - maybe better with rspamd now), but also because on my test runs I saw plenty of legitimate incoming mail that simply would get rejected if I were to enforce DMARC on ingress.
I suspect big players like gmail have some kind of soft enforcement/reputation/machine learning approach to DMARC, but this is way too much work for a small time admin like me.
If someone can solve the spam problem in email without just manually whitelisting specific providers that heavily monitor their users, then they can completely reverse the direction of this problem. Unfortunately it doesn't seem likely someone could do that.
That would work if most spam wasn't from the cartel. I receive single digit percentage of "spam" from randoms, 95%+ is from Gmail/Google, o365 or mailgun/whatever. Not one of them can either filter their users or actually action abuse reports.
The reason for you only seeing spam from the larger providers is that they are filtering out everyone else.
They should also deal with the individual accounts etc, but thats the difficult problem.
I enforce strict spec and behaviour compliance, as well as policies (SPF, dkim, etc). Aside from a very small number of well known spam/abuse sources, it's largely from legitimate mail accounts on Gmail/outlook etc. or idiot msps.
Spam probably isn't as bad as you think it is. It's not the dark old days where 99% of emails are spam.
You should self-host, so you can do the experiment for yourself. You might be surprised. You can always stop self-hosting if you don't like it.
I wish that was true for all cases. My uni email is 99.9% spam and everything comes from official accounts sending irrelevant stuff to everyone and their grandma
Recent estimates show it's down closer to 50% of all emails now. Which is obviously still outrageous.
I self host email and surprisingly do not receive any Spam (apart from the occasional cold calling support@ contact@ etc)
It's not about receiving spam, it's about legitimate messages you send getting filtered as spam.
Wasn't bitcoin supposedly invented to do that?
Ironic given the amount of crypto spam that exists.
It's not in the current providers' interest to help people avoid the current providers' advertisement or avoid paying the providers to host the customers' email.
Seems like AI is in a pretty good position to do that tbh.
You mean filtering? That could get extremely expensive computationally, though. Tools like Rspamd support LLM-based filtering, but traditional Bayesian filtering is vastly cheaper and more computationally efficient. Hm, maybe LLMs can be used as an additional signal for ambiguous messages rather than as the primary filter.
If you are thinking using AI to classify messages, the scale of email spam is so massive that using conventional techniques won't work. It won't make sense energetically speaking IMO.
I think techniques like HashCash by Adam Back should be revisited.
its not really difficult to filter spam off. I spend more time dealing with physical spam in my snailmailbox than my email spamfilter
Self hosting is not that hard actually.
The problem lies with the big e-mail providers marking your e-mail as spam, even if you have everything configured right (MX, SPF, DKIM, DMARC, etc.) and you're not on any blacklists. They use their mysterious opaque algo to determine that your e-mail is spam and you have no idea what you could do to not get in peoples junk boxes.
> They use their mysterious opaque algo to determine that your e-mail is spam and you have no idea what you could do to not get in peoples junk boxes.
And that’s not hard?
Running a modern mail server, a spam filter, a self-hosted webmail, and learning just enough about DNS and networking to get mail through: this is what’s not that hard.
Figuring out how to stay whitelisted, or accept ending up in people’s filters, asking them to whitelist you while your public IP gradually gains trust, is a job.
20 years of running the classic postfix + dovecot stack for about 10 business domains. I’ve recently been testing out stalwart, with just one company domain, which is nice but seems quite fragile, 4 times now it’s killed itself doing some self updates. If it wasn’t for Claude assistance I probably would have given up on it by now. It is quite feature rich though, but its spam detection is not great, too much ham flagged as spam, not enough spam flagged as spam. 2FA is a nice bonus but for some unknown reason it breaks itself every few weeks and I gotta turn it back off again. If I could get stalwart stable I would migrate all my domains to it.
Why switch then? Need more excitement in your life?
I have been running Postfix for 22 years, from a big corporate installation to my own on a Raspberry Pi Zero. It just does its job.
The hardest thing in setting up your emails server is to keep your messages out of spam folders. Even if you are completely legit company that sends out ONLY human-written personal emails, you will get tagged as spam first.
I've self-hosted for about 2 months, and promptly switched to iCloud. Just because I've been spending too much time calling people and telling them to check their spam folders.
Recently our business moved to another domain. We had a guy who has spent over a month "warming up" traffic, so our new domain won't be tagged as spam by default.
And on the other side, projects like stalwart email ( http://stalw.art/ ) and mox ( http://xmox.nl/ ) makes it so much easier to build an e-mailsystem on your server.
I have deployed stalwart as my replacement for mailu and iv never had such an easy time setting up mail, i love it.
Spam ruined email. It's really as simple as that.
I hardly ever see any spam nowadays, happy fastmail user.
Nothing will ever ruin email. A private email server, or a federation of mutually-whitelisting ones, is the ultimate social media, and not at all a bad personal information management system too.
Edit: typo
spam filters are effective at dealing with that
It works both ways right? If you self-host you are seen as the spam.
At this point, self-hosting email is as much a practical convenience as it is a political statement. Or maybe I'm just stubborn.
Interesting topic. I wonder if the AI was given notes to base the article on, or if it was given the data and also made the full analysis itself.
powerful enough models are becoming a reality on personal hardware. Olares is an open-source personal AI cloud OS, supporting local AI, Ollama, Open WebUI, Hermes etc
Hosting my own server with mailinabox. low traffic, but keeping their random subscriptions, newsletters and occasional single purpose address. Haven't had any issues so far.
I switched to M365 as my provider a few years back as my emails kept being flagged as spam.
I've done it for decades, but I don't know much beyond the basics of email hosting tbh, however my understanding is that these days email coming from self-hosted servers carries a much higher spam/junk risk and so often automatically marked as spam.
Maybe there are ways to avoid this happening, but when a few clients all reported that they were not receiving my emails I didn't have much of a choice but to switch.
I would also be interested in self-hosting but have limited understanding of what would be required to ensure my email does not get flagged as spam.
Stopped reading at "The MX record says where the mailbox lives."
Interestingly, DNS does have such a record (MB,) just never widely used - https://datatracker.ietf.org/doc/html/rfc1035/#section-3.3.3
I see no upsides to a Google/Microsoft duopoly in email. Except for those already-far-too-powerful corporations.
This very much depends you been U.S. citizen or not been U.S. citizen.
"PRISM is a code name for a program under which the United States National Security Agency (NSA) collects internet communications from various U.S. internet companies. The program is also known by the SIGAD US-984XN. PRISM collects stored internet communications based on demands made to internet companies such as Google and Apple under Section 702 of the FISA Amendments Act of 2008 to turn over any data that match court-approved search terms."
https://en.wikipedia.org/wiki/PRISM_program
"Section 702 permits the Attorney General and the Director of National Intelligence to jointly authorize targeting of non-US persons reasonably believed to be located outside the United States.
By targeting, it is meant that US persons or persons located in the United States may not be the intended targets of the collection. The targeting must have as its object the acquisition of foreign intelligence information, as this is defined by FISA."
https://en.wikipedia.org/wiki/FISA_Amendments_Act_of_2008
How times change.
I remember when Gmail came around and everyone was more than happy to hand off their email to Google. No spam and ability to send and receive huge (for the time) attachments I think were the main things people wanted.
Not sure if that was before or after everyone was happy to hand off their web browser to Google.
The alternatives to Gmail, at the time, were either Hotmail, or email provided by your ISP who charged money for additional email accounts. Those were the options.
You like using Google and Microsoft products as an end user (even ignoring all the ethical issues with these companies)? That seems surprising to me. I find their products so horrible compared to e.g. Fastmail.
We have to figure out a way to get out of FB social, iOS/Android phone, Google/Microsoft email, AWS/GOOG/MSFT cloud oligopolies. There is no competition and there will be massive rentier capitalism extraction from all these players for the rest of civilization.
There's competition, but no one is interested.
TBH, I blame IT leadership. The easy choice is Google or Microsoft. "No one ever got fired for buying IBM" is now "No one ever got fired for subbing to Office." Someone needs to bite the bullet and stitch together an alternative array of services. They've cultivated their own monoculture by hitting the easy-button and forking over for Office365, etc.
Which if we're being honest, Microsoft and Google products are rarely best-in-class at any one thing that they do. Teams, SharePoint, Google Docs, someone somewhere is doing those things much better. Those competitors just don't necessarily have a whole ecosystem that makes it a big Easy Button for corporate IT to smash. And they sometimes charge as much by themselves as Office365 costs in whole. But I know one problem my own employers IT team struggles with a bit is putting out the constant fire of different teams deciding the Microsoft tool we pay for sucks and going rogue and trying to subscribe to something else. Dropbox was a big one, but Teams and Planner alternatives come up a lot too.
Not pretending to have all the answers but it doesn't seem to be a completely intractable problem.
It starts by just avoiding them as much as possible. E.g. self-host email. If you encounter a system that refuses to receive your email, I give you permission to use your old Gmail account with that system, while using your self hosted account for everything you can.
If your email is Google or Microsoft, there are other people with self-hosted email who feel pressured to use their Google or Microsoft accounts to send email to you.